Talk:Quartz clock

Accuracy
In the article it is written: "Standard-quality resonators of this type are warranted to have a long-term accuracy of about 6 parts per million at 31 °C: that is, a typical quartz wristwatch will gain or lose less than a half second per day at body temperature. If a quartz wristwatch is "rated" by measuring it against an atomic clock's time broadcast, and the wristwatch is worn on one's body to keep its temperature constant, then the corrected time will easily be accurate within 10 seconds per year". I'm a bit puzzled on how half a second per day would eventually add up to 10 seconds a year ... Am i missing something, or someone has lost an order of magnitude in the calculation? jonosphere (talk) 19:40, 12 March 2008 (UTC)
 * The next best thing to a clock that keeps good time is one that has a very predictable error. I think what this was trying to say is that you can *measure* the drift of the wristwatch, and since it can be measured and predicted, you can then correct its reading, with a result that's within 10 seconds a year. If I know my watch gains exactly 17 minutes a week, and I just deduct 17 minutes for every week since I last checked it against WWV, then I don't much care that it's otherwise a lousy watch, as long as its bad behavior is predictable. *Somewhere* I read about a cluster of cheap watches that, when cross-correlated and corrected for their various secular trends, produced a time standard that would otherwise have needed an atomic clock to realize.  Is there a metrologist in the house? --Wtshymanski (talk) 01:43, 31 March 2010 (UTC)


 * The sentence doesn't make sense to me either. What is this "rated" concept?  All except the very cheapest watches are calibrated before sale to obtain the best accuracy with that particular crystal.  I still think that this sentence should be removed unless you can find a reference.  The whole point about cheap watches is that the oscillators are not stable, so will not in general continue to keep correct time (though occasionally you might find one that does).  A good quality quartz watch will keep accurate to within ten seconds a year without any maintenance and without "keeping it warm"!  I recommend that we replace the dubious sentence with mine (which Wtshymanski reverted), or remove it altogether.    D b f i r s   07:05, 31 March 2010 (UTC)
 * A few minutes' work with Google Books under "rating chronometers" will provide suitable references and I'll try to get one ot two. The concept is deeper than just checking the watch against the time beep. It's valid and illustrates what is (evidently) a less-well-known concept of timekeeping. Please don't delete the paragraph in question, it's useful to the encyclopedia. The best thing about Wikipedia is when you get to say "I never heard of *THAT*".  --Wtshymanski (talk) 13:10, 31 March 2010 (UTC)
 * Hmm. Marine chronometer should have had a discussion of this, but mentions nothing after observing the time ball drop in the harbor. And Chronometry is a stub. Horology goes into clocks-as-collectible-objects more than it talks about actually using clocks to accurately measure time. More research is needed! --Wtshymanski (talk) 14:07, 31 March 2010 (UTC)
 * I'm always willing to learn, but the paragraph just doesn't ring true to me. I think that you are confusing "rating" with "regulating", but I'll give you time to come up with a reference. I agree that our aim is to build a better encyclopaedia, so thanks for discussing the matter.   D b f i r s   14:09, 31 March 2010 (UTC)
 * Perhaps I am. I'm not a subject matter expert here. It seems to me that there's two different processes. In the first, you make a clock that keeps time as best you can make it, comparing to, say, astronomical observations. In the second process, you study how the clock varies with temperature and time and use that to correct the time readings made on the clock.  Is the first process "regulating" and the second "rating"?  If you look at the Google on-line copy of Bowditch's "American Practical Navigator", you'll find  a discussion of marine chronometers are rated and the application of the corrected chronometer readings to navigation. While this is getting a little far afield from the observation about cheap quartz watches, the neglect of this topic on the Wikipedia makes me think we should not lightly discard this sentence. I wonder if the original contributor is still around and can expand on this? --Wtshymanski (talk) 16:28, 2 April 2010 (UTC)
 * User:Omegatron contributed a great deal on rating clocks back in May 2005; wonder if that user could point us at some good references? --Wtshymanski (talk) 16:42, 2 April 2010 (UTC)
 * Yes, it would be useful to have an expert on the subject. If I correctly understand the process, quartz crystals are cut to vibrate at a frequency slightly faster than 32,768 Hz, then the crystal is "rated" by comparing its frequency with a standard (e.g. Greenwich time signal or atomic clock), and the regulatory circuitry is adjusted to omit cycles so that the watch or clock keeps exact time.  A good-quality crystal will then remain accurate to within about ten seconds a year, especially if the circuitry also has some temperature compensation built in.  A lower quality crystal will not maintain its stability and will change its frequency at random, or as a result of mild shocks in use.  My first quartz watch (in the early 1970s) cost me several weeks' pay, but it maintained this accuracy for many years until I accidentally threw it across the room, after which I had to regulate it using the adjuster inside the back cover.  I would be interested to learn about self-regulating clocks if you happen to find a good reference.    D b f i r s   18:30, 2 April 2010 (UTC)
 * (outdent) I've got the wrong user listed above - that was just a revert. Most of the text came in 2004; given the half-life of contributors, I think it's unlikely the original contributors are still around. Perhaps the Wikiproject listed above has someone who can explain this. --Wtshymanski (talk) 21:27, 2 April 2010 (UTC)
 * I still think that this doesn't read correctly. The recent addition is more useful.  I'm adding a "citation needed" tag for now in case anyone can make sense of it and provide a reference that explains it.    D b f i r s   07:53, 19 May 2010 (UTC)

What the web thinks of this article
The commercial site Explain that Stuff! has an article on quartz clocks and watches that has a link to this article at the bottom. It characterizes our article as: "This is one of those slightly baffling Wikipedia articles likely to make sense only to people who know enough about the subject to write the article in the first place." Although I wrote some of it, I'm afraid I agree with this assessment. Just thought maybe this would motivate someone to do a rewrite. I'm currently working on other articles, or I'd do it. -- Chetvorno TALK 20:02, 28 April 2010 (UTC)
 * Well, strictly, they link to Quartz oscillator, not *this* article; and that article is written in a considerably different way than things like " if you pass electricity through quartz, it vibrates at a precise frequency " which is close to an explanation but doesn't make any sense if you think about it. (Our proverbial bright 12-year old will know full well that quartz doesn't conduct electricity, for one thing.). --Wtshymanski (talk) 21:09, 28 April 2010 (UTC)
 * Oops! That's embarrassing, I maligned the wrong article.   Yeah, this article is definitely in a higher class than quartz oscillator, although I think the explanation of how quartz clocks work could be improved and the equation for resonant frequency needs to be fixed.  Never mind.  -- Chetvorno TALK 21:54, 28 April 2010 (UTC)


 * Its an interesting article. I'd like to understand it better, but maybe this subject can be spread out among several articles, like a series of lessons.   I collect watches.  I prefer automatic movements, to battery-operated watches, but this article gives me a newfound respect for quartz watches.  — Preceding unsigned comment added by 206.192.35.125 (talk) 14:00, 11 April 2012 (UTC)

Accuracy versus temperature
Can someone add a section on accuracy versus temperature? If you take your watch off at night, it will cool to room temperature. What happens then? Woz2 (talk) 03:15, 5 July 2010 (UTC)
 * Most clocks and watches have some form of temperature compensation, but I don't know how it is implemented. Do we have an expert on quartz oscillator technology who could add a paragraph, or someone with time to research this?  I've noticed a very slight effect on some watches when they are cooled close to freezing, but the compensation circuitry seems very efficient at the accuracies I need (a few seconds per month).    D b f i r s   07:01, 5 July 2010 (UTC)
 * From what I've read, it seems to me that temperature compensation is only implemented in high end quartz watches. I still have to find a good reference on this though. An accuracy of a few seconds per month could be achieved just by keeping the watch in contact with the body all day, and at room temperature at night, given a typical temperature coefficient of 0.035 ppm/K². --Edgar.bonet (talk) 16:37, 9 November 2011 (UTC)
 * You might be correct, but then how do really cheap watches achieve the same accuracy under widely varying temperatures? (Mine seem to do so, but I suppose this is original research).  I agree that the first generation of cheap watches probably had no temperature compensation.  I remember being surprised that my first (high end) quartz watch kept the same amazing accuracy at near-freezing as it did at near-body temperature.   D b f i r s   16:56, 9 November 2011 (UTC)

Well, yes, looking at your own clock may be personal research, but interesting anyway if you can put numbers on your observations. The expected average drift rate for an uncompensated quartz is
 * $$a\left((\langle T \rangle - T_0)^2 + \sigma_T^2\right) + d_0$$

where $$a \approx -0.035\;\mathrm{ppm/K^2}$$ is the temperature coefficient (I found this value on an actual spec), $$\langle T \rangle$$ is the average temperature, $$T_0 \approx 25^\circ C$$ the turnover temperature, $$\sigma_T^2$$ the temperature variance, and $$d_0$$ the drift rate at the turnover temperature (this drift may be purposefully introduced in order to compensate for the expected value of the first term). Are your observations on cheap watches consistent with this formula?

I found a few resources that more or less claim that, unlike high end watches, “ordinary” quartz watches lack temperature compensation:
 * Prometheus Bound – The Final Paradigm of Horological Evolution, by Carlos Perez
 * In Pursuit of Perfection: Thermocompensated Quartz Watches and Their Movements, by Bruce Reding
 * Introduction to Quartz Frequency Standards, by John R. Vig (page "Static Frequency versus Temperature Stability")
 * Quartz Crystal Resonators and Oscillators For Frequency Control and Timing Applications - A Tutorial, by John R. Vig (PPT, expanded version of the previous, see notes on slide 134) — Preceding unsigned comment added by Edgar.bonet (talk • contribs) 15:21, 10 November 2011 (UTC)

A stupid question
This might be a stupid question: A battery-operated, Quartz watch still has gears, doesn't it? I guess the battery powers a small transmitter. The transmitter generates a signal. the signal passes through the quartz material, and passing thru the quartz adjusts the timing of the signal to thus adjust the timing of the motor that keeps the gears turning. Is that a good "idiots guide to quartz watches" explanation? marc s.. dania fl 206.192.35.125 (talk) 14:05, 11 April 2012 (UTC)


 * Errr... no! The battery powers an electronic circuit that drives the clock. This circuit keeps the quartz oscillating by applying some feedback to it. The circuit also has a counter that counts the oscillations. Every time the count reaches some fixed number (typically 32768 for a typical watch quartz) this driving circuit sends an electric pulse to a stepper motor. The motor, through some gears, moves the hands forward by one second. The electric pulse sent to the motor does not travel through the quartz. --Edgar.bonet (talk) 20:01, 11 April 2012 (UTC)
 * NB: The explanation above is for the “standard” analog quartz watch. But then there are watches with digital displays and no gears, and there is the Spring Drive with a quartz time base, gears, hands, but no stepper motor. --Edgar.bonet (talk) 20:24, 11 April 2012 (UTC)

Broken link
The link on reference 9 is broken. 76.126.244.254 (talk) 05:57, 13 March 2015 (UTC)